Peptides capable of binding specific target molecules have been useful for a variety of research and product development work in the biological, medical, and pharmaceutical fields. Typically in these fields, the target molecules have been those of biological significance, such as peptide epitopes of diseased or pathogenic cells, surface receptor proteins, signaling proteins, proteins involved in the etiology of biological responses such as immune T cell and B cell responses, and targets that are useful for biological research, for example, proteins that bind to nucleic acids or other proteins of biological interest, to which a specific binding partner is needed for identification and screening purposes.
The most longstanding tool for obtaining peptides with specific target binding activities has come from exploitation of the immune response, originally by obtaining polyclonal antibodies, such as IgGs and IgMs induced in an animal challenged by injection of the target molecule to obtain a B cell response and the isolation of antibodies from blood. Subsequently, monoclonal antibodies were obtained from clones of hybridoma cells that produce a single species of antibody that binds a specific epitope of a target molecule. Useful antibodies have the ability to bind target molecules with a molar dissociation constant (Kd) typically of 10−7 M or less and more typically 10−8 M to 10−9 M. These technologies relied entirely on exploiting the natural biological immune system that is capable of recombining the coding sequences for the hypervariable domains of antibodies to create an enormous diversity of antibodies and that has the ability to naturally stimulate the propagation of the few that bind to the target molecule of interest. In the case of monoclonal antibodies, the naturally induced propagation was replaced by the human made ability to select and culture clones that express the specific antibodies.
While use of antibodies has proven to be a powerful tool for obtaining peptides that bind to specific target molecules, antibodies have limitations in utility for industrial applications. First, they must be made from whole blood, chicken eggs, or from hybridoma tissue cultures—all of which are expensive and low yield production systems in comparison to what would be needed for industrial scale binding of target molecules. Second, they rely on the three dimensional structure of the binding domain formed in the hypervariable region of the antibody molecule, requiring production of a rather large protein (even in the case a single chain variable fragments) to obtain one molar binding equivalent for the relatively smaller peptide domain needed to bind each mole of target molecule.
A more recent alternative to antibodies for the identification of target binding domains was the development of bacterial protein display systems, most notably phage display systems. These systems display peptides or whole protein sequences as fusions with surface proteins, e.g., with phage display the peptides are expressed as fusions with phage particle proteins.
Phage display is one of the most powerful and well established technologies for exploring the sequence space of combinatorial random peptide libraries. Typically, foreign proteins/peptides are expressed on the surface of M13 bacteriophage as fusions to either the minor coat protein pIII or the major coat protein pVIII. Libraries with a diversity of 107-1011 peptides (having 5-40 residues) can be screened against the target molecule (immobilized on beads or adsorbed in microtiter wells) and enriched for specific binders through iterative rounds of binding and infection. Binding of individual clones from enriched pools are detected by ELISA and the amino acid sequence of the binding peptide deciphered by sequencing the DNA in the phage particle. The amino acid sequence of the peptide ligands may then be compared to protein databases to identify potential endogenous interacting proteins in silico.
Phage display technology has been primarily useful for identifying protein-protein, protein-peptide and protein-DNA interactions and as such, has been particularly useful as a research tool to identify target peptides that interact with physiologically important proteins, e.g., antibodies and receptors, or that bind to specific sequences of DNA or to potential drug candidates in order to discover potential physiological target proteins for pharmaceutical applications. Phage display, however, has rarely been used to identify peptides that bind to small molecules of industrial importance, such as contaminants in a processing stream or metal ions. While useful as a research tool, phage display has not been shown to be practical as tool for actual production of peptide domains for commercial deployment in any industrial process or product.
A handful of metal binding peptides, however, have been described. One metal binding motif found in metal binding peptides is a sequence of 6 histidines (“polyhistidine”), which is known to be capable of binding nickel. Expression vectors are available that contain nucleic acid sequences with promoters linked to regions encoding polyhistidine containing peptides to make so called “his-tagged” fusion proteins that can quickly be isolated from a cellular extract by relying on the ability of the polyhistidine domain to bind to nickel immobilized on a column. The bound protein is eluted from the column using an excess of imidazole. The polyhistidine binding domain can be cleaved from the eluted protein with a protease if the cloning vector additionally encodes a protease substrate site in-frame with the fusion protein.
The most well known polyhistidine binding domain is a peptide comprising a core sequence of six histidines, as exemplified by the sequence YSHHHHHHLAGTA (SEQ ID NO: 1), which has a molar dissociation constant for nickel of 2.3×10−11 M. Another known polyhistidine binding domain is a 12 amino acid peptide that is a 6 mer repeat of a histidine-glutamine dipeptide, i.e., (HQ)6 (SEQ ID NO:2), Arginine has also been implicated as an important contributing amino acid residue for nickel binding because the consensus sequence RHXHHR (SEQ ID NO:3), where X is most frequently histidine, was also shown to bind nickel with high affinity (Jie et al., Chemical Biology & Drug Design (2006) 68:107-112). Jie et al. identified that sequence by screening peptides form a bacterial library engineered to display proteins on flagella and suggested that bacteria displaying such a sequence might be useful as a biologically derived waste water remediation agent.
Using a similar system, a very different motif for a peptide that binds metal was disclosed by Behnaz et al. (Iranian Journal of Biotechnology (2005) 3:180-185), which showed that the cysteine rich peptide GCGCPCGCG (SEQ ID NO:4) displayed on the surface of E. coli via a fibrinea fusion protein was capable of binding metals in the relative order lead>cadmium>nickel. Regarding cysteine, E. coli that displayed on its surface by fusion to the OmpX membrane protein the cysteine containing peptide LCCYWSYSRMCKN (SEQ ID NO:5) (which was selected from a library of randomly generated 11-mer with two cysteines separated by seven amino acids and each flanked by two amino acids) was shown to bind gold particles in suspension (Kaviani, Biological Applications NNIN REU (2006) Research Accomplishments, p 12-13). A proline/hydroxy containing gold binding peptide of the sequence LKAHLPPSRLPS (SEQ ID NO:6) was identified by phage display using M13 (Nam et al. Science (2006) 312:885-888). Similarly, but in unrelated work, several hydroxyl rich peptides identified by phage surface display, but having no particular consensus sequence, were shown to bind aluminum, with the peptide VPSSGPQDTRTIT (SEQ ID NO:7) showing particularly strong binding (Zao et al., Appl. Microb. and Biotech. (2005) 68:505-509). Gold binding peptides have been suggested as potentially useful in the assembly of nanostructures for microelectronics. Others have shown that metal binding peptides displayed on M13 phage libraries could be useful for biotemplating catalysts to improve catalytic activity (Nelner et al., ACS Nano (2010) 4:3227-3235).
Despite the suggestions, phage display or other bacterial display systems alone are not suitable for deployment in practical industrial scale processes such as water remediation, recovery of precious metals or removal of contaminants from processing streams. This is because the quantity of binding sites needed to bind target molecules from industrial scale processing streams is extremely large. By way of example, a typical phage titer produced via bacterial cell culture is on the order of 1012 particles per mL. Even assuming each particle displayed 103 binding proteins, each binding one equivalent of a target molecule, it would require 6.02×108 mL, or 602,000 liters of cell culture to produce enough particles just to bind one mole of a target molecule. One mole of nickel is 60 grams of material and a typical water intensive industrial production process, such as processing corn in a 250,000 bushel per day wet mill facility, which uses hundreds of thousands of liters of water per hour, can extract as much as 6 lbs (2700 grams) of nickel in just a three hour period. Therefore, to use the nickel binding domains expressed on phage particles to bind all the nickel produced in one day from a com wet mill facility, or other agricultural processes that generate large volumes of waste water that contain extracted metals such as nickel, e.g., soybean processing, may use about 218 million liters of phage culture to produce enough particles. Such large scale production is commercially impractical. Display on bacterial surfaces is even less practical because the total number of displayed molecules per bacterium is about the same as phage, but the maximum titer of bacteria is on the order of 109 cells per mL. Therefore, it would require at least 1000 times the amount of bacterial culture to display enough binding sites on flagella or fibrinea as would be required to display the same on a phage particle.
There is a practical, industrial need for specifically binding small molecules from industrial processing streams that would be useful for water remediation, removal of contaminants from food products, and large scale purification of naturally occurring small molecules.